ABSTRACT
In the study of disease spatial distribution it is often appropriate to ask questions related to the local properties of the relative risk surface rather than models of relative risk per se. Relative risk estimation concerns the "global" smoothing of risk and estimation of true underlying risk level, whereas cluster detection is focused on local features of the risk surface where elevations of risk or depressions of risk occur. Global clustering basically assumes that the risk surface is clustered or has areas of like elevated risk. An uncorrelated surface, should display random changes in risk with changes in location and so should be much more variable in risk level and have few contiguous areas of like risk. The chapter considers three different scenarios for clustering: single region hot spot relative risk detection, clusters as objects or groupings, and clusters defined as residuals.
